INDEX

Explanations

references to rabbits and bunny-related terms

oai_token-act-pair · gpt-3.5-turbo

instances of the words "rabbit" and "bunny."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

omething

-0.93

ician

-0.85

igmat

-0.84

inia

-0.78

rylic

-0.78

orie

-0.77

itutional

-0.76

ructure

-0.75

xit

-0.73

inen

-0.73

POSITIVE LOGITS

MQ

1.18

 Hole

0.96

 rabbit

0.88

meat

0.86

 Rabbit

0.85

 Wilde

0.82

 Nest

0.79

 rabbits

0.76

 Hunt

0.72

wald

0.69

Activations Density 0.021%

references to rabbits and bunny-related terms

instances of the words "rabbit" and "bunny."

No Comments

No Known Activations